Green innovation in the Yangtze River Delta is closely related to higher-quality integrated development, and knowledge diversity is crucial to the realization of regional green technology innovation and development. This study measured the green innovation efficiency of cities in the Yangtze River Delta region from 2010 to 2018 utilizing the Super-SBM model based on undesired outputs. In addition, this study used patent data to investigate regional knowledge deversity, including related variety, and unrelated variety, and to examine the spatio-temporal characteristics of green innovation efficiency and the threshold effect of knowledge diversity. The results demonstrated that related variety was positively correlated with the efficiency of urban green innovation, which was in line with extant studies. Unrelated variety was accompanied by an increase in urban science and technology investment and expansion of urban scale, and the negative effect of knowledge unrelated variety was significantly weakened. This study deepened the understanding of the mechanism of action of diversity, which is conducive to the sustainable development goals as regards the formulation of policies related to green innovation and the development of various types of cities.

Innovative development to some extent overcomes the path-dependent characteristics of knowledge proximity, and the diversity and complexity of regional knowledge and technology foster regional economic growth. This study constructs a dynamic mechanism for maximizing the enterprise profit to explore the conditions enabling regions that can achieve knowledge accumulation and technological leapfrogging. Utilizing the fitness landscape methodology based on customs data from 30 provinces in China and employing genetic algorithms, we analyze two modes of enterprises with similar and different initial conditions under varying leapfrogging scenarios. By applying profit maximization rules within the fitness landscape of the product space, we examine how to promote regional industrial transformation and upgrade. The results indicate that a larger leap range in the production of products by enterprises, reflecting stronger innovative capabilities, significantly enhances the region’s ability for industrial transformation and upgrade. There is a strong positive relationship between the initial industrial foundation and regional economic development. When regions share a similar industrial base, the complexity of regional knowledge becomes more critical. Therefore, an increase in the foundational development of industries and technology-intensive products can enhance the potential for regional path innovation. Past research has primarily been based on the path dependence theory, emphasizing that the most crucial aspect for underdeveloped economies is the accumulation and development of foundational capabilities. However, our study contends that, for emerging economies, it is crucial to enhance the diversity and complexity of knowledge and technology. Particularly, it is essential to, based on a comprehensive understanding of their own strengths, open up new areas and new arenas in development, continually shaping new growth drivers and new strengths. This emerges as a significant factor in regional growth and leapfrog development.

The concept of the “space of flows” emphasizes cross-space resource sharing arising from element flows in the digital age and has become an important topic in digital economy (DE) research. The DE is a key driver of global economic growth, whereas carbon emissions (CE) remain a major barrier to achieving global sustainable development goals. Existing research primarily examines the DE-CE relationship from a static perspective based on regional attributes. However, there is a lack of in-depth analysis on the digital economy spatial correlation network (DESCN) and its impact on CE from the perspective of the space of flows. This study aims to address this academic gap within the Chinese context. Our study applies the modified gravity model and social network analysis (SNA) method to characterize the DESCN, and further uses fuzzy set qualitative comparative analysis (fsQCA) to explore the network’s influence on CE. The key findings are as follows: (1) The DESCN exhibits multi-threaded and densifying characteristics, with increasingly frequent inter-regional cooperation, closer internal connections, and a continuously expanding overall scope. (2) While the DESCN exerts a significant overall impact on CE, its specific effects vary significantly. Simply elevating a region’s status within the network is not enough to achieve regional carbon reduction and may even result in increased emissions. (3) To fully unlock the carbon reduction potential of the DE, regions must consider other factors, including population size, energy consumption, industrial structure, and openness.